Chapter 20 - Avant-garde techniques in machine for detecting financial fraud in healthcare

Abstract

Machine learning is a blooming technology in the current Internet world. Mostly in all the fields, the machine learning started exploring its techniques. One such area is the medical field. In this sector, the patients' records should be precise for the medication process. Machine learning techniques help the medical field in this regard. In this prominent medical sector, there is a possibility of collecting huge amounts of data. These data include the financial details like electronic payments, credit card details, pin numbers, etc. This will lead to a challenge to secure these details from the fraudulent utilizations. Whatever security methods followed, the hackers are updating themselves to get the electronic payment details. Thus, there should be a perpetual refinement for the design. Some scenarios like virus attack, whaling, smishing, etc., which have a great impact on financial losses. This process gets the important sensitive details about the card holders, and this makes the threat for the electronic card users. To overcome this issue, a proficient technology must be used to diagnose such sensitive issues which were conducted in the electronic payment cards such as credit cards. In this work, different types of algorithms like random forest, K-nearest neighbor (KNN), Naive Bayes, and logistic regression are used to detect the fraudulent usage of the credit cards. Here, different types of machine learning approaches are used. Following are some of the speculations which are considered in this work. In recent number of transactions, the total amount of transaction statistics, and statistics based on the regional area. With the help of these details the trusted owner and the fraud use of the card can be found out easily. The accuracy of the data can be calculated by the dataset, which is collected from the real time entries. The comparative analysis generates the better results. This is proved visually with the analysis done in this work.